LabVIEW

cancel
Showing results for 
Search instead for 
Did you mean: 

FPGA compile fails with Single-Cycle Times Loops

My FPGA VI compiles without error. If I add Single-Cycle Timed Loops (SCTLs) to it and try to compile, LabVIEW crashes while generating intermediate files, at step 4 of 6.
 
I'm running LabVIEW 8.5.0 under Windows XP on a Dell M6300 with an Intel Centrino Duo.
 
After experimenting a bit, I created a minimal VI that replicates the problem.
 
The VI is simply an SCTL containing a case structure. Each case writes to the same boolean global variable, and my guess is that this is the problem.
 
Am I not allowed to write to the same global more than once in an SCTL, even if a case structure guarantees that only one write would actually occur?
 
The enclosed VI demonstrates this, although I didn't enclose the globals file, so just create one and link the two globals in the case structure to the same boolean global.
 
Thanks.
0 Kudos
Message 1 of 4
(3,185 Views)
Hello,

Shared resources like Local or global variables require arbitration. You can configure the arbitration options for each resource interface of FPGA I/O, FIFO, and memory items if you need to optimize the FPGA VI. Other resources that include arbitration, such as controls and global variables, include fixed arbitration settings. You cannot change the arbitration option for such resources, so in some situations, you might need to modify the code on the block diagram to avoid resource contention.

Info:
Code Generation Error 61056: Invalid Arbitration for Single-Cycle Timed Loop
Managing Shared Resources (FPGA Module)

Best regards,
Nick_CH
0 Kudos
Message 2 of 4
(3,169 Views)
With further experimentation, I'm getting the impression that if I write to a global in an SCTL, I'm not allowed to write to that global anywhere else in the VI, even if it's guaranteed that the writes cannot occur simultaneously.
 
But my problem isn't that- it's that LabVIEW crashes when I try this, without telling me why.
0 Kudos
Message 3 of 4
(3,153 Views)

Hi RonW,

I can see how LabVIEW crashing without a reason would be frustrating.  This was reported to R&D (# 58593) for further investigation.  I appreciate the feedback and apologize for the inconvenience. 

Jennifer R.
National Instruments
Applications Engineer
0 Kudos
Message 4 of 4
(3,135 Views)